Supplementary information Supplementary figures Supplementary Figure 1. Chromosomal synteny between zebra finch (left) and great tit (right) genome assemblies. Different colours represent different chromosomes. The genome comparison was done by using the software LASTZ1. 1 Supplementary Figure 2. Patterns of accelerated evolution of the EGR1 gene in the tit lineage. Upper Panel: Pairwise dn/ds rates between 46 bird genomes2 (for a full list of species see Supplementary Data 6 of gene EGR1). Increased dN/dS value for the great tit-ground tit comparison is indicated with an arrow. Lower panel: Variable sites in the EGR1 coding region for the 29 birds, reference bird and ground tit EGR1 represented as a conservation sequence logo plot3. The sequence of the reference bird and ground tit is given below, and the major variant is highlighted in blue. Altogether 53 positions are variable, 10 are fixed differences between ground tit and great tit EGR1 and none is specific to the reference bird. 2 Supplementary Figure 3. CpG methylation patterns across genomic features. CpG methylation profile of (a) CpG islands and shores, (b) CpG islands and shores within gene bodies and up to 10kb upstream of transcription start sites, and (c) gene bodies, including 10kb upstream and downstream. Shaded areas denote variances. Vertical lines denote boundaries of (a, b) CpG islands and (c) gene bodies (annotated transcription start site to transcription termination site). 3 Supplementary Figure 4. Expression and methylation profile of sweep and non sweep genes. Neuronal expression, CpG, and non-CpG methylation profile of sweep and non-sweep genes with (a) no expression, (b) fragments per kilobase of exon per million fragments mapped (FPKM) between 0 - 1, (c) FPKM between 1 - 10, and (d) FPKM between 10 - 100. (e) Expression profile of sweep and non-sweep genes. For example, genes with a FPKM of 10 are expressed at 10 times the rate of genes with a FPKM of 1. 4 Supplementary Figure 5. DNA methylation levels are associated with evolutionary rates. (a) dN/dS (nonsynonymous substitution rate devided by synonymous substitution rate), (b) dN (nonsynonymous substitution rate), and (c) dS (synonymous substitution rate) values of the lower (lowly methylated) and upper (highly methylated) quartiles of genes in the brain. Four methylation categories with following sample sizes were considered: CpG gene body methylation (1712 lowly and 1618 highly methylated genes), CpG methylation at transcription start sites (2023 versus 2040), non-CpG gene body methylation (972 versus 918) and non-CpG metahylation at transcription start sites (1673 versus 1686). Significance was addressed with a one-sided Mann- Whitney-U-Test). *** P<0.001, n.s. = not significant; the box in a box plot represents the range between upper and lower quartiles, the solid horizontal line denotes the median, and the whiskers show the most extreme data point, which is no more than 1.5 times the length of the box away from the box.. 5 Marker order Supplementary Figure 6. Comparison of marker order between the genome assembly and two linkage maps. Each point represents a SNP that was on the NL and UK linkage maps and which could be placed on the assembly scaffold by BLAT. Orange symbols represent the Netherlands map and blue symbols the UK map. Spearman rank correlation coefficients (Rho) between assembly order and linkage map order are indicated and were typically >0.99. X-axis is marker order (units are marker number from 1 to n, where n is number of markers) on the genome assembly. Y-axis is position in cM on the linkage map. Supplementary Figure 7. Great tit genome annotation pipeline. Supplementary Figure 8. Chromosome wide average of nucleotide diversity estimated for synonymous sites and all sites as a function of chromosome size. The solid lines denote local regressions. Supplementary Figure 9. Folded site frequency spectra (SFS) derived for regions of chromosomes 1, 5 and 20. Variaabbility denotes the proportion of polymorphic sites and the neutral model denotes the expected site frequency under the neutral theory. Supplementary Figure 10. Diversity and rearrangements of great tit Z chromosome. (a) Watterson’s Θ (a measurement of nucleotide diversity) measured in a sliding window of 10 kB for the Z-chromosome. Diversity for 221 male birds was obtained from the ANGSD pipeline. (b) Chromosome synteny of the zebra finch Z chromosome and the great tit Z chromosome illustrating substantial intrachromosomal rearrangements since the split of the two species. Supplementary Methods Study species Great tits are common, widely distributed, cavity nesting passerines. The range of this super-species spans Europe across Asia to the Pacific, including populations in the Middle East, and North West Africa. From this distribution area four species groups have been identified: major (whole Europe, Siberia and northwest Africa), cinereus (from Iran to India and southeast Asia), minor (China, Japan and eastern Russia) and bokharensis (central Asia) 4,5. Most studies have concentrated on the northwest European Parus major subspecies and our sequenced bird belongs to this subspecies. The major subspecies has very distinctive black-yellow-white plumage. The males have brighter colouration and they are larger than females. The great tits living in north-west Europe experience variable environmental conditions which affects their ecology. Great tits are insectivorous during the summer and when the number of insects drops down during the winter they partly switch to seeds and berries. In addition they are regarded as a partial migrant species and during the breeding season they form monogamous pairs and they are territorial. After the breeding season great tits live in flocks which can even include other species. In their natural habitat they rely on cavities in hardwood trees for nesting. However, they also readily breed in artificial nest boxes and this makes it possible for detailed life history data to be collected for entire populations. This has also allowed them to colonise human environments where nest boxes, other human created cavities, and supplemental feeding support populations. As a result of their abundance and ready use of nest boxes, great tit populations have been intensively studied across their range; some ongoing studies have now collected data on tens of thousands of individuals spanning 60 years. As a result they are now one of the most widely studied species in ecology and evolution. Genome sequencing and assembly Sample preparation The great tit used for this study originated from a captive population artificially selected for four generations for avian personality6. The parental generation was taken from two natural populations as 10- day old nestlings and hand reared until independence in the lab. The birds were selected on the basis of their performance in two behavioural tests. Pairs were formed in such a way that there was no inbreeding in the fourth generation. Eggs were laid in an aviary of 2 x 4 x 2.5 m and when females started incubating the whole clutch was transferred to a foster nest in a natural population. The offspring hatching from these eggs then were transported to the lab at day 10 after hatching and hand reared. The reference great tit was anesthetized using Isoflurane and medical oxygen and euthanized by subtracting all blood from the carotid artery under protocol number CTE-0705 Adendum I, from the Animal Experiment Committee from the Royal Netherlands Academy of Sciences (DEC-KNAW). DNA was isolated from blood samples using the Puregene system (Gentra, USA). The bird was subsequently dissected and all organs were stored separately and snap-frozen at -80 and stored in RNA later. RNA was extracted from eight tissues (bone marrow, homogenized half of the brain, breast filet, higher intestine, kidney, liver, lung and testis) and RNA isolation was performed on 25 mg thawed tissue segments using the miRNeasy mini kit from Qiagen, following the protocol of the manufacturer. For library preparation we used the standard Illumina TruSeqRNAseq protocol using a different barcoded adapters (Illumina) for each of the samples. De Novo RNA assembly using Trinity The combined 1 billion reads from all eight tissues were simultaneously de novo assembled using the Trinity software package7 version r2013-02-25. RNAseq data from the different tissues were combined for the following reasons (1) This would result in one set of predicted gene models with uniform transcript names across the data set and (2) this would also allow better assembly of transcripts present at low levels in multiple tissues. Because of the high depth of the RNAseq data, we first normalized the data. We noticed that due to the high coverage, the data contains a substantial number of retained introns, resulting in an unrealistically large number of alternative transcripts and unrealistically large average size of the transcripts. We therefore extensively explored different settings for min (2-40) and max (100-1000 (or unlimit)) to find the optimum settings for these parameters to exclude retained intronswhile minimizing the exclusion of rare transcripts. The most optimal settings for our data and the settings used for the final de novo assembly were: --min_kmer_cov 15; --max_cov 400 and --PARALLEL_STATS. Following the normalization, the Trinity assembly was subsequently run using the Trinity default settings. Using these settings, we obtained a total of 101,289 assembled transcripts ranging in size from 201 to 16,061 bp and with an average size of 1,335 bp. Reference based RNA assembly using Tophat/Cufflinks The normalized RNAseq data was also analysed with Tophat version v2.0.108 (Bowtie v2.1.09) using the following settings: number of hits allowed: 20 (default); inner mate distance 50bp, std inner mate distance 150bp, optimizing mapping accuracy with the "--read-realign-edit-dist" option. Subsequently the tophat alignment output was analysed with Cufflinks version 2.2.010 including the following settings: --overlap- radius 5 ('distance to split genes' - to avoid merging of adjacent genes; the default is 50 but we observed that this occasionally led to merging of genes that are close together).